RESEARCH ARTICLE

RAB-10 Regulates Dendritic Branching by Balancing Dendritic Transport Caitlin A. Taylor, Jing Yan, Audrey S. Howell, Xintong Dong, Kang Shen* Howard Hughes Medical Institute, Department of Biology, Stanford University, Stanford, California, United States of America * [email protected]

Abstract

OPEN ACCESS Citation: Taylor CA, Yan J, Howell AS, Dong X, Shen K (2015) RAB-10 Regulates Dendritic Branching by Balancing Dendritic Transport. PLoS Genet 11(12): e1005695. doi:10.1371/journal.pgen.1005695

The construction of a large dendritic arbor requires robust growth and the precise delivery of membrane and protein cargoes to specific subcellular regions of the developing dendrite. How the microtubule-based vesicular trafficking and sorting systems are regulated to distribute these dendritic development factors throughout the dendrite is not well understood. Here we identify the small GTPase RAB-10 and the exocyst complex as critical regulators of dendrite morphogenesis and patterning in the C. elegans sensory neuron PVD. In rab-10 mutants, PVD dendritic branches are reduced in the posterior region of the cell but are excessive in the distal anterior region of the cell. We also demonstrate that the dendritic branch distribution within PVD depends on the balance between the molecular motors kinesin-1/UNC-116 and dynein, and we propose that RAB-10 regulates dendrite morphology by balancing the activity of these motors to appropriately distribute branching factors, including the transmembrane receptor DMA-1.

Editor: Andrew D. Chisholm, University of California San Diego, UNITED STATES Received: March 26, 2015 Accepted: October 31, 2015 Published: December 3, 2015 Copyright: © 2015 Taylor et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Funding: This work was funded by Howard Hughes Medical Institute Funding, 9/1/2013–8/31/2018, KS, Investigator, http://www.hhmi.org/. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist.

Author Summary Building a complex dendritic arbor requires tremendous cellular growth, and how membrane and protein components are transported to support a rapidly growing, polarized dendrite remains unclear. We have identified the small GTPase RAB-10 as a key regulator of this process, providing insight into both dendritic development and the control of trafficking by small GTPases.

Introduction There is great diversity in the structure and complexity of dendritic arbors across neuron types, and establishing the correct dendritic morphology is critical for the proper connectivity and function of neural circuits. A developing dendritic arbor must target a specific receptive field, adopt the appropriate neuron-specific architecture, and avoid overlapping in connectivity with itself and neighboring dendrites. A number of extrinsic cues and intrinsic mechanisms help orchestrate the formation of these complex neuronal morphologies, including transcriptional programs, extracellular guidance cues, and contact-dependent repulsive molecules that mediate self-avoidance [1–5].

PLOS Genetics | DOI:10.1371/journal.pgen.1005695

December 3, 2015

1 / 24

RAB-10 Regulates Dendritic Branching by Balancing Dendritic Transport

Dendritic arbor development requires tremendous cellular growth and likely has specialized membrane trafficking demands. Little is known about how the transport of branching factors and membrane components is coordinated across a large, polarized neuron. Dendrites are more sensitive than axons in their reliance on the membrane supply from secretory pathways [2,6], and they have distinct transport needs. For example, a set of dendritic arbor reduction (dar) genes have been identified which are required for dendritic arbor outgrowth but not axon growth. These dar genes are important for ER-to-Golgi transport [6]. In addition, the Rab GTPases, a conserved family of small GTPase proteins that regulate membrane identity and vesicle trafficking [7–9], are likely important for the polarization and outgrowth of neurites, though their precise role in both axonal and dendritic development remains unclear [10]. One of these small GTPases, Rab10, has been shown to mediate membrane trafficking in several polarized cell types, including neurons [11–18]. The importance of Rab10 for endosomal sorting and endocytic recycling has been demonstrated in Drosophila epithelial cells [15] as well as in C. elegans neurons [16–17] and intestinal epithelial cells [11–14]. In hippocampal neurons, Rab10 is required for directional membrane trafficking in the growing axon [19–20], and it has been shown to directly bind the anterograde motor kinesin1 via the adaptor protein JIP1 [21]. In addition to sorting and trafficking cargo to the appropriate destinations, a growing neurite must appropriately dock and release membrane and protein cargoes. The docking of Rab10-postive vesicles is required for axon outgrowth in hippocampal neurons [22]. Rab10 has also been linked to the exocyst, a secretory complex responsible for polarized exocytosis [23– 25]. The exocyst is an octameric complex that was initially identified in yeast for its role in membrane addition during bud outgrowth, and it has since been shown to have conserved functions in polarized growth across several cell types [25–28]. In Drosophila, the sec5 component of the exocyst complex is required to establish polarity in the oocyte [26], and the sec15 component is also required for the polarized delivery of photoreceptors [28]. Both Rab10 and the exocyst complex are required for branch outgrowth in the Drosophila trachea [25]. Taken together, these data suggest that both the exocyst complex and RAB-10 support outgrowth in polarized cells; however, the roles of RAB-10 and the exocyst have never been examined in a growing dendrite. In Caenorhabditis elegans, the PVD sensory neuron forms elaborate dendritic arbors that are organized into orthogonal tiers of characteristic menorah-shaped units [29–30]. The PVD neuron has proved to be useful in the study of the molecular mechanisms of dendrite morphogenesis. Importantly, the dendritic branches of PVD maintain self-avoidance [31], a key feature across neural networks. Extracellular cues are essential for PVD dendritic morphogenesis: the branching receptor DMA-1, expressed in PVD, instructs spatially-restricted growth of dendritic branches via interaction with the SAX-7/MNR-1 adhesion complex [32–34]. Previous work has also identified various intracellular processes important for PVD dendritic morphology, including transcription factors that instruct cell fate [29,35], microtubule-associated proteins [36], and membrane fusion proteins that prevent over-branching [37]. However, the role of polarized membrane trafficking in the growing PVD dendrite remains uncharacterized. In this work, we identified a novel role for the small GTPase RAB-10 in patterning the dendritic arbor of PVD. We demonstrate that both RAB-10 and the exocyst complex are required for proper outgrowth and patterning of the PVD dendritic arbor. RAB-10 and EXOC-8 are also required for the appropriate localization of branching factors such as DMA-1. Additionally, we found that the molecular motors kinesin and dynein are important for anterior-posterior patterning of the PVD dendritic arbor. The PVD dendritic outgrowth phenotype was also observed in a recently published report [38], and our work further identifies a role for RAB-10 in dendritic patterning beyond its role in promoting outgrowth. We propose that RAB-10

PLOS Genetics | DOI:10.1371/journal.pgen.1005695

December 3, 2015

2 / 24

RAB-10 Regulates Dendritic Branching by Balancing Dendritic Transport

regulates dendrite morphogenesis by balancing anterograde and retrograde transport via molecular motors.

Results wy787 disrupts dendritic arbor morphology We visualized the morphology of the PVD neurons by expressing membrane-bound-GFP under the control of a cell-specific promoter (ser2prom3:myrGFP) (Fig 1A and 1B) [29]. In wild-type animals, one anterior and one posterior primary dendritic branch extend from the cell body during the L2 larval stage. Dense menorahs consisting of a series of orthogonal 2°, 3°, and 4° branches emerge from the anterior and posterior 1° dendrites as the animal progresses through subsequent larval stages (L2-L4) (Fig 1A, 1B and 1E). To understand the molecular mechanisms of dendritic branching, we performed a forward genetic screen to identify mutations that affect the dendritic morphology of PVD. From this screen, we isolated individuals with the fully penetrant and recessive mutant allele wy787, which show a severe reduction in the number of dendritic branches (Fig 1C and 1D). wy787 mutant animals also have several non-neuronal phenotypes, including intestinal vacuoles and reduced brood size. In wy787 mutants, the number of secondary branches is reduced from 42 ± 6 in wild-type to 28 ± 9 in wy787, and the number of quaternary branches is reduced from 114 ± 15 to 17 ± 10 (S1 Table). In addition to the reduction in branch number, we noticed a shift in branch distribution. In wy787 mutants, considerably more branches are found in the distal anterior region of the 1° dendrite, where wild-type animals typically grow few branches (Fig 1B and 1D). To quantify the distribution of menorah, we divided the entire PVD dendritic arbor into four regions: the region posterior to the cell body (designated as “-1”) and three equal-length regions in the anterior dendrite (designated as “+1, +2 and +3”). We defined a “branch complexity” index to characterize the completeness of menorahs in each region. This branch complexity metric weights three characteristics equally: (1) the number of secondary branches, (2) the percentage of secondary branches that form a tertiary branch, (3) and the average number of quaternary branches per tertiary branch (Fig 1G). In wy787 mutants, branch complexity is severely reduced in the +2, +1, and -1 regions but increased in the +3 region. (Fig 1C, 1D and 1H). The severe decrease in branch complexity in the +1 and -1 regions is the result of a near-total loss of 3° and 4° branches in the +1 and -1 regions (S1 Table). The increase in branch complexity in the anterior +3 region led us to hypothesize that the branching defects cannot be entirely explained by an overall lack of branching activity, but are instead the result of abnormal distribution of branching activity along the anterior-posterior axis. The lack of branches in the posterior regions (+1 and -1) could be due to a defect in 2° branch outgrowth or stabilization. We distinguished between these possibilities by examining dendrite morphology during earlier developmental time points. In the L2 and L3 larval stages, wild-type and wy787 showed no significant difference in the number of secondary filopodia (Fig 1F), which suggests that wy787 could be defective in branch-stabilization. However, in wy787 the posterior 2° branches are often shorter than they are in wild-type and often do not grow in the correct orthogonal orientation. These defective branches cannot reach the tertiary branch-point (Fig 1C), a region enriched for the branch-promoting complex, MNR-1/SAX-7, that supports tertiary branch outgrowth [33]. By the L4 stage, wy787 had significantly fewer 2° branches, and those 2° branches that reached the tertiary line in the posterior regions (+1 and -1) failed to form 3° branches (Fig 1D and 1F). This developmental failure is consistent with a defect in both stabilization and outgrowth of 2° branches as well as both initiation and maintenance defects in 3° and 4° branches. This defective branching in the posterior regions is reminiscent of the phenotype of the branching receptor mutant dma-1 [32–33].

PLOS Genetics | DOI:10.1371/journal.pgen.1005695

December 3, 2015

3 / 24

RAB-10 Regulates Dendritic Branching by Balancing Dendritic Transport

Fig 1. rab-10(wy787) causes defects in PVD dendritic morphology. (A, B) PVD was labeled with a cell-specific marker (ser2prom3::myrGFP). The dendritic arbor covers the entire body of a wild-type animal. 2° and 3° branches are established by the L3 larval stage, and by the L4 larval stage the dendritic arbor has fully elaborated with 4° branches. A second cell body appears in some micrographs because ser2prom3::myrGFP also labels the non-branching PDE neuron. (C, D) wy787 mutants have severely disrupted dendritic morphology. 2° branches emerge in the L3 stage but most do not reach the 3° branchpoint or extend 3° branches. Full menorah form only in the distal anterior region of the dendrite. (E) Schematic of the PVD dendritic arbor in wild-type and wy787 animals. (F) Quantification of 2° branch filopodia in larval stages in wild-type and wy787 animals. (G) Definition of branch complexity index used to quantify menorah completeness per region for each genotype. (H) Quantification of subcellular distribution of branch complexity in wild-type and wy787 animals. Scale bars represent 20 μm. Error bars represent SEM. * p

RAB-10 Regulates Dendritic Branching by Balancing Dendritic Transport.

The construction of a large dendritic arbor requires robust growth and the precise delivery of membrane and protein cargoes to specific subcellular re...
NAN Sizes 1 Downloads 8 Views